Click here to log on to the Wikpedia on-line Encyclopedia

 

(But first read the following two reviews of the Wikpedia on-line Encyclopedia:)


By Michael Cook
Saturday, 17 December 2005

A defamatory entry in the on-line encyclopaedia Wikipedia has put one of the internet’s great success stories under the spotlight.

One of the most remarkable success stories of the internet has been the rise and rise of Wikipedia, the on-line encyclopaedia. For decades, the gold standard for reference books has been the weighty tomes of the Encyclopaedia Britannica, with its two-volume index and hundreds of scholarly contributors. Wikipedia threatens the eminence of such works because it is free, universally available and up-to-date. Nowadays articles in Wikipedia are regularly consulted by journalists, students and even academics. The founder of the site, Jimmy Wales, describes it as “an effort to create and distribute a free encyclopaedia of the highest possible quality to every single person on the planet in their own language.”

Nothing if ambitious, but Wikipedia’s history justifies it. It began only four years ago in 2001 and has since grown to 2,550,000 articles, including more than 860,000 in the English-language version. About 1500 articles were added every day in October this year. The German, French, Japanese, Polish, Italian, Swedish and Dutch, all have more than 100,000 articles. Enthusiasts are attempting to build up versions in other languages, including Pidgin English, Scots and Esperanto. Good luck to them!

To have paid experts to write these millions of articles would have been impossible. Wikipedia paid nothing. Instead, in a stunning example of the power of the internet to recruit dedicated manpower, it marshalls thousands of anonymous volunteers, aka “Wikipedians”. Any reader can alter the text of an article or even create a new one, although registered users have more authority and privileges in the editing process. This makes Wikipedia the most visible achievement of the open source software movement. With enough volunteers each bringing his grain of sand, it proves that an industry-competitive product can be created. Wikipedia has done for knowledge what Linux did for operating systems.

But like all software, Wikipedia has vulnerabilities. The value of its information can be corrupted by inaccuracy, bias –– or vandalism. This weakness was painfully exposed this month when a former editorial page writer for USA Today discovered that he had been defamed in his Wikipedia biography. To his dismay John Seigenthaler, 78, a former aide to Robert Kennedy and a pall-bearer at his funeral, read that he might have been involved in the assassination of the Kennedy brothers. “I have no idea whose sick mind conceived the false, malicious ‘‘biography’’ that appeared under my name for 132 days on Wikipedia,” he fumed.(1)

It turned out that the entry had been a prank by a courier manager living near Seigenthaler.(2) “I didn’t think twice about just leaving it there because I didn’t think anyone would ever take it seriously for more than a few seconds,” the remorseful vandal said.(3) This and another scandal prompted Wikipedia to tighten up its editing rules this week.

For fans of traditional print-based encyclopaedias with professional editing teams this highly publicised and embarrassing incident confirmed all of their misgivings. Robert McHenry, a former editor of the Britannica, argues that it is “fatally flawed”. Even apart from malicious vandalism, he says, Wikipedia articles inevitably slide into mediocrity because the editors themselves are often ill-informed or cannot express themselves clearly. “The user who visits Wikipedia to learn about some subject, to confirm some matter of fact, is rather in the position of a visitor to a public restroom. It may be obviously dirty, so that he knows to exercise great care, or it may seem fairly clean, so that he may be lulled into a false sense of security. What he certainly does not know is who has used the facilities before him,” he wrote last year. (4)

Wikipedians brush off inadequacies of their darling with the argument that the more the entries are used, the more accurate they become. In the language of open source geeks, “Given enough eyeballs, all errors are shallow.” Of course, this happens only if an article is popular enough. In practice Wikipedia is like a Third World road system. You can zoom along sign-posted superhighways, but as soon as you veer off onto country roads, there are potholes and mysterious byways, some of which end in magnificent mansions, while others peter out into gravel lanes and featureless scrub.

An example, I searched for an entry on James McAuley, a major Australian poet. As fans of Australian poetry (yes, it exists!) know well, McAuley was one of a pair of hoaxers who created a poet named Ern Malley, wrote his poems, and got them published in an avant-garde poetry magazine. It generated the greatest controversy in Australian literary history. Well, Ern — who never existed —has several well-written pages devoted to him, the work of an ardent devotee, no doubt, but McAuley — who did exist — has none.

Not all the science entries are terrific either, although science and technology are supposed to be strong points. Despite the fact that therapeutic cloning has been on the front page of newspapers since the birth of Wikipedia, its entry contains merely 12 brief sentences. And these contain several minor errors — major errors if you feel that therapeutic cloning presents major moral issues.

The depth and accuracy of the articles broadly reflect the interests of people who use the internet. First of all, there are the geeks. The article on Star Trek, for instance, is quite comprehensive and the one on Vulcans is about three times as long as Ern Malley’s. Ern at least wrote some real poems, but Vulcans never existed at all, although it is hard to detect from reading the article. Broadband internet access has a comprehensive and very useful article, as do most internet and computing terms.

And then there are the people who are obsessed by current affairs. The article on Hurricane Katrina is endless, covering nearly every aspect of the disaster. In fact, the more recent the news, the longer and more detailed the article is likely to be. The life and death of Stanley Tookie Williams III, the leader of the Los Angeles Crips gang, is covered at great length, including his execution this week. For journalists, Wikipedia has become an invaluable resource, which may explain the good press it receives.

In fact, the balance of expert opinion is tilting inexorably in favour of Wikipedia. The leading science journal Nature defended it this week as a “free, high-quality global resource”.(5) It even took the trouble to conduct what it described as “an expert-led investigation” in which reviewers compared 42 science entries in Wikipedia with the corresponding entry in the Britannica. The result was about even: four serious errors in each, although Wikipedia had more smaller errors.(6) It was a nasty blow to Britannica’s prestige.

Perhaps the dispute is really a matter of perceptions. Wikipedia is being promoted as an encyclopedia. In fact, it is something altogether different. It will never, ever, be a conventional encyclopaedia and it is useless to expect it to rival the thoroughness, accuracy and balance of the Britannica. It is a warehouse of facts, not a school of wisdom.

The Britannica is a characteristic child of the Enlightenment rationalism with its dream of mastering all possible knowledge. Indeed, Alastair MacIntyre, in his luminous study of contemporary moral philosophy, Three Rival Views of Moral Inquiry, used the 19th century Britannica in its ninth edition as the epitome of modern (that is, pre-post-modern) rationalism. It represents the view that a dispassionate, objective, educated intellect can order, dominate and possess all of truth.

Wikipedia is the creation of a post-modern age. It is far more modest in its claims than the Britannica, as befits post-modern scepticism about truth. It depends upon the wisdom of crowds: that truth is an evolving concept defined by the interaction of many minds. The same philosophy underlies the search algorithms of Google and the voting system for "American Idol". It assumes that a good article is basically a sandcastle of grains of fact heaped higher and higher. So long as the facts can be empirically verified and the castle stays standing, the article is a good one. But Wikipedia has no way of assessing whether the sandcastle is beautiful, or big enough, or too big, or whether it was worth building at all. Wikipedia can aspire to excellence in articles about technology, or current affairs, or biography, but it will almost always begrossly inadequate when dealing with ideas.

So Wikipedia can never achieve more than a close approximation to truth, especially non-empirical truth –– but near enough is good enough for most post-moderns. In fact, this might even help to explain why Wikipedians are so dismissive of their critics.

Anyhow, there is no turning back. This week Wikipedia – which did not exist five years ago — was the 34th most visited site on the internet. It has enormous deficiencies, but also opportunities. The glaring gaps in its coverage are an opportunity for people to contribute articles with their own values and make them available to all other users of the internet. And by doing so, they can actually improve on the Britannica.

Here’s my favourite example. Wikipedia’s entry on ethics is superficial, short, unbalanced and unsophisticated. In short, awful. What’s to be done? Well, you can turn to the Encyclopaedia Britannica where you can read an objective, polished, comprehensive article written by an anonymous contributor. But the Britannica's notion of Olympian objectivity is something of a myth. who turns out to be none other than the notorious Professor Peter Singer, a supporter of infanticide and euthanasia? Maybe not. Maybe the solution is to persuade your favourite philosopher to rewrite Wikipedia’’s entry.

----------------------------------------------------------

Michael Cook is editor of MercatorNet. Disclosure: an entry for “MercatorNet” on Wikipedia was deleted as “Vanity/advertising” by a Wikipedian overlord named TitoXD.

Notes

(1) John Seigenthaler. “A false Wikipedia ‘biography’’”. USA Today. Nov 21, 2005.
(2) “Tracking down the Wikipedia prankster”. ZDNet News. Dec 15, 2005.
(3) “Author apologizes for fake Wikipedia biography”. USA Today. Dec 11, 2005.
(4) Robert McHenry. “The Faith-Based Encyclopedia”. TechCentralStation. Nov 15, 2004.
(5) “Wiki’s wild world.” Nature. Dec 15, 2005.
(6) “Internet encyclopaedias go head to head.” Nature. Dec 15, 2005.

--------------------------------------------------------------------------------------------------------------------------------------------------------------

Wikipedia: garbage in, garbage out   

By John Bambenek 
John Bambenek is a columnist and freelance writer who blogs at Part-Time Pundit. His biography was deleted from Wikipedia by its editors, but at one stage it falsely listed him as a child sex offender for over an hour and a half.
 
Thursday, 04 January 2007

Not every internet pundit is a fan of the world's biggest encyclopaedia.

Jim Wales, the founder of Wikipedia, describes Wikipedia as "an effort to create and distribute a multi-lingual free encyclopedia of the highest possible quality to every single person on the planet in their own language." The motto is "the free encyclopedia that anyone can edit."

However, Wikipedia has several drawbacks that call into question its legitimacy as a serious service. One of these drawbacks lies in the claim that it is a site that "anyone can edit." The problem with this model is that not everyone can edit and even a smaller portion would edit. For instance, people without computers cannot edit. Business people and executives are generally far too busy to edit. That leaves a small subset of people.

That subset of people is technophiles who are generally white, generally male and generally not experts in what they are editing. In fact, one of the original people involved with Wikipedia has criticized what he calls "anti-elitism" in Wikipedia. What this means is that there is a general distrust for "experts" and a preference for editors who are less knowledgeable. As a friend put it, "they can code Perl so they think they're experts in evolutionary biology."

Wikipedia recognizes this problem and responds with the maxim "out of mediocrity, excellence". The idea that enough mediocrity can eventually produce excellence is novel and interesting, but unconvincing. A better and perhaps more accurate mantra might be "garbage in, garbage out."

Disregarding the cognoscenti in favor of the incognizanti leads to very little verification of facts in Wikipedia. Every article that is created on Wikipedia is generally reviewed for quick compliance with policies (ie, not vandalism, not slanderous) -- essentially to see if it passes the "smell test". If the page looks complete and looks like it contains real information, it stays.

Take, for instance, the case of Climbing Jack which was an April Fool's joke article that remained on Wikipedia for over eight months before being removed. Another more damning example is the case of Erdosville, Nebraska. Since the town does not exist, it would have been a snap to verify that it did not exist. The entry stayed up for over a year, and interestingly enough, a real estate agent advertised being able to sell houses near "world-famous Erdosville."

Wikipedia also has an interesting test for what gets to be including in articles called "verifiability". This concept is summarized by Wikipedia as "the threshold for inclusion is verifiability, not truth". Verifiability can be further amended as something that can be verified over the web. There is a bizarre Internet-centric view of information, namely, that if it isn't on the Internet, it does not exist.

The lack of input validation into Wikipedia leads not only to outright false information but a variety of articles that have no content as well. Wikipedia brags that it has over 1.5 million articles in the English edition (the one used for the rest of this article). Many of those articles, however, are stubs. Stubs are what Wikipedia calls articles that only have a sentence or two of content and act as placeholders for a topic yet to be written. According to Wikipedia's own estimates there are over 765,000 stubs. This estimate could off by about 20 per cent because Wikipedia rounded up in its counts and articles might be listed as a stub in multiple categories. This leaves about 600,000 stubs. That means over two in five articles in Wikipedia are  stubs. That does not include articles that are stubs that have not been tagged by someone. This is in line with other estimates.

Wikipedia also has a policy forbidding the use of original research in Wikipedia. If it hasn’t been published somewhere else, it cannot be used. However, over 180,000 articles are tagged by Wikipedia as not having necessary sources. That is, over 12 per cent of the articles in Wikipedia contain assertions which have not been documented.

Lastly, there are pages that exist solely as "disambiguation" pages. These are pages listing the many alternatives for a certain word. For instance, if you search Wikipedia for "George Bush", you get a page that lists the alternatives of what you might mean. There are about 73,000 disambiguation pages. There are also about 32,000 articles that are nothing but lists of other articles.

Adding the numbers above shows that 58 per cent of articles on Wikipedia have no intellectual merit whatsoever. In addition, one study found that between 1 and 2 per cent of Wikipedia articles have been plagiarized in part or in whole. Wikipedia's own estimates list that only 34 per cent of articles are more than 2 kilobytes (300 words). Articles under that size are likely stubs or articles with no real content. This is all before looking at articles that are on useless or inconsequential subjects such as almost every sex position or piece of sexual slang.

Once an article or piece of information makes it into Wikipedia, the presumption is that the article or item remains. This has lead to a dramatic increase of stubs, for instance, but also of capricious deletion policies. The notion of "notability" is in the eye of the beholder. Hence, there are many non-notable articles that are in Wikipedia because they've "always been there". On the flip side, articles have been deleted simply because someone is upset at the author, subject, or article.

Most article deletion votes take place with 10 or fewer people voting. In December, Wikipedia boasted 17,000 active editors. That means with 1 in 1,700 voting, "consensus" is achieved. Or more accurately, if there is a small group of cranky editors who want to cause problems, they can get their way.

Due to the selection bias of authors and editors in Wikipedia, there is a bizarre sense of priority in what is included and expanded in Wikipedia. The article on Britney Spears is 13 pages long, the same length as the article on Henry the VIII. The rule on notability in Wikipedia suggests that articles included should have relevance to people 100 years from now. It is hard to imagine that Britney Spears will even be remembered in 100 years. King Henry the VIII, on the other hand, did found the Anglican Church and split from Rome.

Pythagoras, the father of numbers, gets only six pages on Wikipedia. That is the same length as the article on Leeroy Jenkins who had a briefly popular five minute web video on World of Warcraft -- and Leeroy Jenkins isn't even his real name. Pythagoras only gave us the Pythagorean Theorem. Bubb Rubb even gets an article and all he did was get interviewed by a local news show and shout "whoooo!" into the camera.

Almost every game created for the Nintendo Entertainment System (the first one from 1985) has an article on Wikipedia. Runescape, a marginally popular online game, has 45 pages. Pokémon has over 226 pages dedicated to every aspect of it. By way of contrast, the Book of Genesis entry on Wikipedia takes up only 64 pages. Something is profoundly absurd about an encyclopaedia entry on a video game being over three times longer than the entry for the Book of Genesis.

Wikipedia notability constitutes whatever is popular, if only for one day, to technophiles. For instance, there are 174 articles on Battlestar Galactica. This includes articles not only on each of the actors, but articles on each of the characters, technology, and even the religion portrayed in the show. Granted the show is popular and very entertaining (even to me), but it is not likely to be remembered 10 years from now.

Another problem with Wikipedia is that it will not censor for the protection of minors. One could argue for the inclusion of a relevant image in an article on genital warts. However, it is purely pornographic to include an image of full-frontal nudity in an article on indecent exposure. There are galleries of pornographic images that are not used in any article. School children use Wikipedia for research and there are caches of porn contained in it. Wikipedia will do nothing about the issue for fear of "censorship."

Rules do exist on Wikipedia to provide the appearance of objective decision-making. However, one of the rules that Wikipedia has is that there are no rules. If someone with enough clout decides that a certain action will "improve Wikipedia", all other rules are jettisoned. For instance, creating or editing a biography on oneself is against the rules. Nevertheless, Jim Wales edited his own biography.

Wikipedia has gained tremendous popularity in the few years of its existence. In such time, the weaknesses of the model have become apparent. Some of these stem from the model they've chosen for the system, but no small part of the problem is the bias of its editors. There are ludicrous emphases on some subjects of no consequence, a startling number of articles with no content whatsoever, and a policy system that can be overruled at the whim of an administrator. Many teachers and librarians have written against Wikipedia's usefulness as a primary reference. The fact is that Wikipedia is untrustworthy as anything other than a quick place to look to find other sites with reliable information.

--------------------------------------------------------------------------------------------------------------------------------------------------------